A population based, individually matched case control study is proposed to examine serum levels of vitamins A and E, selenium and cholesterol in relation to lung cancer risk. Sera from individuals who participated in a 1974 serum collection study and who subsequently developed lung cancer will be compared with those of matched controls from the 1974 study for levels of these micronutrients. Lung cancer cases will be identified from the Washington County cancer register. Cases and controls or appropriate surrogates will be interviewed to ascertain usual frequency of eating foods high in fats, vitamin A (including carotene) and vitamin E. The relationship between serum levels and past dietary intake of the 2 vitamins will be determined. Demographic, socioeconomic and personal characteristics will be studied and their effects, if any, controlled in the analysis.